Perfusion distribution among lung microvessels is thought to be controlled mainly by arteriolar constriction and dilation. However, effects of this activity on microvessel perfusability are unknown. Furthermore, considerable anatomic evidence suggests that active perfusion control is possible within the lung microvessels themselves, although this subject has never been studied. We will address these deficits in our understanding of pulmonary microvascular control using methods we developed. We will identify vasoactive agents that act directly on pulmonary microvessels versus those that act on pulmonary arterioles or venules. We will also determine how these agents affect red cell microvessel perfusability. We will conduct these studies in normal and hypoxic lungs, and in those injured by sepsis to determine the clinical relevance of this control. We propose the following specific aims: 1. Vasoactive response of lung microvessels to pharmacologic agents that produce whole-lung vasoconstriction. We will vasoconstrict lungs using angiotensin-II, bradykinin, serotonin, the thromboxane analog U46619, or hypoxia. Latex particles of a specific diameter (1.0, 2.0, 3.0, or 4.0 ?m) will be infused into each lung during vasoconstriction and the lungs will then be rapidly frozen. Particle densities within the microvessels will be measured in confocal histological images for particles of each diameter (one diameter per lung). These diameter-specific particle densities will be used to quantify the average microvessel diameter in lungs of each treatment group. Vasoconstrictors that produce microvessel diameters similar to those in matched-flow controls will be assumed to affect mainly arterioles. Vasoconstrictors that produce microvessel diameters smaller than those in matched-flow controls will be assumed to affect mainly microvessels. Vasoconstrictors that produce microvessel diameters larger than those in matched-flow controls will be assumed to affect mainly venules. These results will allow us to identify the vascular segment in which each agent exerts the majority of its vasoconstriction: in arterioles, microvessels, or venules. We will also examine the effects of hypoxia, and the effects of Rho kinase and nitric oxide synthase (NOS), in normal and hypoxic lungs to further clarify the pharmacologic reactivity of each vascular segment. The studies in this aim provide the baseline data for aims 2 and 3. 2. Pulmonary microvessel red cell perfusability response to pharmacologic agents that produce whole-lung vasoconstriction. The studies in this Aim quantify the ability of red blood cells to flo through microvessels under the conditions utilized in Aim 1. The goals of this Aim are to determine the clinical relevance of the microvessel diameters identified in Aim 1. Our objective is to learn if the changes in microvessel diameters identified in Aim 1 translate to corresponding changes in the red cell perfusability. 3. Vasoactive response of lung microvessels to pharmacologic agents known to produce whole-lung vasoconstriction in lungs injured by sepsis. The goals of this aim are to learn how lung microvessel diameters and reactivity are affected by clinically relevant lung injury. Perfusion distribution among lung capillaries is known to be markedly disturbed by sepsis, and to cause ventilation/perfusion abnormalities. However, sepsis- induced changes in microvessel diameters and vasoreactivity may be responsible for this as well. This is a subject about which nothing is known, and our studies will address it for the first time. We will use the latex particle and red cell methods employed in Aims 1 and 2 to determine how microvessel diameters and red cell perfusion are affected by sepsis (LPS infusion), and to also determine how the pharmacologic responsiveness of each vessel segment is altered by sepsis. Results of our studies will expand our basic understanding of pulmonary microvascular flow regulation in normal and injured lungs, and lead to new treatments that improve lung capillary perfusion in lung injury.